The use of immunoglobulins as therapeutic agents has increased dramatically in recent years and have expanded to different areas of medical treatments. Such uses include treatment of agammaglobulinemia and hypogammaglobulinemia, as immunosuppressive agents for treating autoimmune diseases and graft-vs.-host (GVH) diseases, the treatment of lymphoid malignancies, and passive immunotherapies for the treatment of various systemic and infectious diseases. Also, immunoglobulins are useful as in vivo diagnostic tools, for example, in diagnostic imaging procedures.
One critical issue in these therapies is the persistence of immunoglobulins in the circulation. The rate of immunoglobulin clearance directly affects the amount and frequency of dosage of the immunoglobulin. Increased dosage and frequency of dosage may cause adverse effects in the patient and also increase medical costs.
IgG is the most prevalent immunoglobulin class in humans and other mammals and is utilized in various types of immunotherapies and diagnostic procedures. The mechanism of IgG catabolism in the circulation has been elucidated through studies related to the transfer of passive immunity from mother to fetus/neonate through the placenta or yolk sac or through colostrum (maternofetal transfer of IgG via transcytosis) in rodents (Brambell, Lancet, ii:1087-1093, 1966; Rodewald, J. Cell Biol., 71:666-670, 1976; Morris et al., In: Antigen Absorption by the Gut, pp. 3-22, 1978, University Park Press, Baltimore; Jones et al., J. Clin. Invest., 51:2916-2927, 1972).
The involvement of certain receptors in the maternofetal transmission of maternal IgGs was first suggested by Brambell's group in their study on the intestinal absorption of maternal antibodies from ingested milk in newborn rats (Halliday, Proc. R. Soc. B., 143:408-413, 1955; Halliday, Proc. R. Soc. B., 144:427-430, 1955; Halliday, Proc. R. Soc. B., 148:92-103, 1957; Morris, Proc. R. Soc. B., 148:84-91, 1957; Brambell et al., Proc. R. Soc. B., 149:1-11, 1958; Morris, Proc. R. Soc. B., 160:276-292, 1964). Brambell et al. suggested, based on the observation that heterologous IgGs interfered with the transmission of a specific antibody, that IgG molecules from various species might have sufficiently similar structures or sequences that bind to common receptors (Brambell et al., Proc. R. Soc. B., 149:1-11, 1958).
A high-affinity Fc receptor, FcRn, has been implicated in this transfer mechanism. The FcRn receptor has been isolated from duodenal epithelial brush borders of suckling rats (Rodewald et al., J. Cell Biol., 99:154s-164s, 1984; Simister et al., Eur. J. Immunol. 15:733-738, 1985) and the corresponding gene has been cloned (Simister et al., Nature, 337:184, 1989 and Cold Spring Harbor Symp. Quant. Biol., LIV, 571-580, 1989). The later clonings of FcRn-encoding genes from mice (Ahouse et al., J. Immunol., 151:6076-6088, 1993) and humans (Story et al., J. Exp. Med., 180:2377-2381, 1994) demonstrate high homology of these sequences to the rat FcRn, suggesting a similar mechanism of maternofetal transmission of IgGs involving FcRn in these species.
Meanwhile, a mechanism for IgG catabolism was also proposed by Brambell's group (Brambell et al., Nature, 203:1352-1355, 1964; Brambell, Lancet, ii:1087-1093, 1966). They proposed that a proportion of IgG molecules in the circulation are bound by certain cellular receptors (i.e., FcRn), which are saturable, whereby the IgGs are protected from degradation and eventually recycled into the circulation; on the other hand, IgGs which are not bound by the receptors are degraded. The proposed mechanism was consistent with the IgG catabolism observed in hypergammaglobulinemic or hypogammaglobulinemic patients. Furthermore, based on his studies as well as others (see, e.g., Spiegelberg et al., J. Exp. Med., 121:323-338, 1965; Edelman et al., Proc. Natl. Acad. Sci. USA, 63:78-85, 1969), Brambell also suggested that the mechanisms involved in maternofetal transfer of IgG and catabolism of IgG may be either the same or, at least, very closely related (Brambell, Lancet, ii:1087-1093, 1966). Indeed, it was later reported that a mutation in the Fc-hinge fragment caused concomitant changes in catabolism, maternofetal transfer, neonatal transcytosis, and, particularly, binding to FcRn (Ghetie et al., Immunology Today, 18(12):592-598, 1997).
These observations suggested that portions of the IgG constant domain control IgG metabolism, including the rate of IgG degradation in the serum through interactions with FcRn. Indeed, increased binding affinity for FcRn increased the serum half-life of the molecule (Kim et al., Eur. J. Immunol., 24:2429-2434, 1994; Popov et al., Mol. Immunol., 33:493-502, 1996; Ghetie et al., Eur. J. Immunol., 26:690-696, 1996; Junghans et al., Proc. Natl. Acad. Sci. USA, 93:5512-5516, 1996; Israel et al, Immunol. 89:573-578, 1996).
Various site-specific mutagenesis experiments in the Fc region of mouse IgGs have led to identification of certain critical amino acid residues involved in the interaction between IgG and FcRn (Kim et al., Eur. J. Immunol. 24:2429-2434, 1994; Medesan et al., Eur. J. Immunol. 26:2533, 1996; Medesan et al., J. Immunol., 158:2211-2217, 1997). These studies and sequence comparison studies found that isoleucine at position 253, histidine at position 310, and histidine at position 435 (according to Kabat numbering, Kabat et al., In: Sequences of Proteins of Immunological Interest, US Department of Health and Human Services, 1991, which is hereby incorporated by reference in its entirety), are highly conserved in human and rodent IgGs, suggesting their importance in IgG-FcRn binding.
Additionally, various publications describe methods for obtaining physiologically active molecules whose half-lives are modified either by introducing an FcRn-binding polypeptide into the molecules (WO 97/43316; U.S. Pat. Nos. 5,869,046; 5,747,035; WO 96/32478; WO 91/14438) or by fusing the molecules with antibodies whose FcRn-binding affinities are preserved but affinities for other Fc receptors have been greatly reduced (WO 99/43713) or fusing with FcRn binding domains of antibodies (WO 00/09560; U.S. Pat. No. 4,703,039). However, none of these publications disclose specific mutants in the IgG constant domain that affect half-life.
Prior studies have demonstrated that certain constant domain mutations actually reduce binding to FcRn and, thereby, reduce the IgG in vivo half-life. PCT publication WO 93/22332 (by Ward et al.) discloses various recombinant mouse IgGs whose in vivo half-lives are reduced by mutations between about residue 253 and about residue 434. Particularly, substitutions of isoleucine at position 253; histidine at position 310; glutamine at position 311; His at position 433; and asparagine at position 434 were found to reduce IgG half-life.
Modulation of IgG molecules by amino acid substitution, addition, or deletion to increase or reduce affinity for FcRn is also disclosed in WO 98/23289; however, the publication does not list any specific mutants that exhibit either longer or shorter in vivo half-lives.
In fact, only one mutant of mouse IgG1 that actually exhibited increased half-life, the triple mutation Thr252 to Ala, Thr254 to Ser, and Thr256 to Phe, has been identified (WO 97/34631).
In view of the pharmaceutical importance of increasing the in vivo half-lives of immunoglobulins and other bioactive molecules, there is a need to develop modified IgGs and FcRn-binding fragments thereof. (particularly modified human IgGs) that confer increased in vivo half-life on immunoglobulins and other bioactive molecules.